transport mechanism(s) of poly (amidoamine) dendrimers across caco-2 cell monolayers

7
International Journal of Pharmaceutics 265 (2003) 151–157 Note Transport mechanism(s) of poly (amidoamine) dendrimers across Caco-2 cell monolayers M. El-Sayed a,1 , C.A. Rhodes b , M. Ginski b,2 , H. Ghandehari a,a Department of Pharmaceutical Sciences, School of Pharmacy, University of Maryland, 20 N. Pine Street, Baltimore, MD 21201, USA b Department of Pharmaceutics, Guilford Pharmaceuticals Inc., 6611 Tributary Street, Baltimore, MD 21224, USA Received 14 January 2003; received in revised form 6 May 2003; accepted 7 July 2003 Abstract The objective of this research was to investigate the mechanism(s) of transport of generation 2 (G2) poly (amidoamine) dendrimers across Caco-2 cell monolayers. The contribution of an energy-dependent process such as adsorptive endocytosis was investigated by determining G2 permeability at 4 and 37 C. The contribution of P-gp efflux to transport was examined by determining the apical to basolateral (AB) and basolateral to apical (BA) permeability of 14 C-paclitaxel in presence of G2, and by determining AB and BA permeability of G2 in presence of paclitaxel. The permeability of G2 and 14 C-mannitol was investigated in the presence of palmitoyl carnitine to determine the contribution of the paracellular pathway. Permeability of G2 at 4 C was significantly (P< 0.05) lower than that observed at 37 C. AB and BA permeability of 14 C-paclitaxel did not change in the presence of G2. AB and BA permeability of G2 did not change in the presence of paclitaxel. The permeability of G2 and 14 C-mannitol increased significantly (P< 0.05) in the presence of palmitoyl carnitine, and in addition, 14 C-mannitol permeability was increased in presence of G2. The permeability of G2 across Caco-2 cell monolayers appears to involve a combination of paracellular transport and an energy-dependent process, possibly adsorptive endocytosis. G2 dendrimers do not appear to be substrates for the P-gp efflux system. © 2003 Elsevier B.V. All rights reserved. Keywords: Drug delivery; Oral permeability; Poly (amidoamine) dendrimers Among water-soluble polymers that have shown potential as oral drug carriers is a family of cas- cade polymers named poly (amidoamine), PAMAM, dendrimers (Wiwattanapatapee et al., 1998, 2000; Corresponding author. Tel.: +1-410-706-8650; fax: +1-410-706-0346. E-mail address: [email protected] (H. Ghandehari). 1 Present address: Department of Bioengineering, University of Washington, P.O. Box 352255, AERL 341A, Seattle, WA 98195, USA. 2 Present address: Shire Laboratories Inc., Department of Prefor- mulation Sciences, 1550 East Gude Drive, Rockville, MD 20850, USA. El-Sayed et al., 2002, 2003). Our previous studies on the transport of cationic PAMAM-NH 2 den- drimers of generations 0–4 (G0–G4) showed that the smaller dendrimers, G0–G2, exhibited mod- erate permeability across Caco-2 cell monolayers (1 × 10 6 –10 × 10 6 cm/s) (El-Sayed et al., 2002). Of these three dendrimers, G2 appeared to be the most promising candidate for oral delivery due to its higher permeability, as well as its higher number of surface amine groups that can be used for immobilization of drug molecules. The objective of this research was an initial attempt to investigate the mechanism(s) of G2 transport across Caco-2 cell monolayers. 0378-5173/$ – see front matter © 2003 Elsevier B.V. All rights reserved. doi:10.1016/S0378-5173(03)00391-0

Upload: m-el-sayed

Post on 04-Jul-2016

214 views

Category:

Documents


1 download

TRANSCRIPT

Page 1: Transport mechanism(s) of poly (amidoamine) dendrimers across Caco-2 cell monolayers

International Journal of Pharmaceutics 265 (2003) 151–157

Note

Transport mechanism(s) of poly (amidoamine) dendrimersacross Caco-2 cell monolayers

M. El-Sayeda,1, C.A. Rhodesb, M. Ginskib,2, H. Ghandeharia,∗a Department of Pharmaceutical Sciences, School of Pharmacy, University of Maryland, 20 N. Pine Street, Baltimore, MD 21201, USA

b Department of Pharmaceutics, Guilford Pharmaceuticals Inc., 6611 Tributary Street, Baltimore, MD 21224, USA

Received 14 January 2003; received in revised form 6 May 2003; accepted 7 July 2003

Abstract

The objective of this research was to investigate the mechanism(s) of transport of generation 2 (G2) poly (amidoamine)dendrimers across Caco-2 cell monolayers. The contribution of an energy-dependent process such as adsorptive endocytosiswas investigated by determining G2 permeability at 4 and 37◦C. The contribution of P-gp efflux to transport was examinedby determining the apical to basolateral (AB) and basolateral to apical (BA) permeability of14C-paclitaxel in presence of G2,and by determining AB and BA permeability of G2 in presence of paclitaxel. The permeability of G2 and14C-mannitol wasinvestigated in the presence of palmitoyl carnitine to determine the contribution of the paracellular pathway. Permeability ofG2 at 4◦C was significantly (P < 0.05) lower than that observed at 37◦C. AB and BA permeability of14C-paclitaxel did notchange in the presence of G2. AB and BA permeability of G2 did not change in the presence of paclitaxel. The permeability ofG2 and14C-mannitol increased significantly (P < 0.05) in the presence of palmitoyl carnitine, and in addition,14C-mannitolpermeability was increased in presence of G2. The permeability of G2 across Caco-2 cell monolayers appears to involve acombination of paracellular transport and an energy-dependent process, possibly adsorptive endocytosis. G2 dendrimers do notappear to be substrates for the P-gp efflux system.© 2003 Elsevier B.V. All rights reserved.

Keywords: Drug delivery; Oral permeability; Poly (amidoamine) dendrimers

Among water-soluble polymers that have shownpotential as oral drug carriers is a family of cas-cade polymers named poly (amidoamine), PAMAM,dendrimers (Wiwattanapatapee et al., 1998, 2000;

∗ Corresponding author. Tel.:+1-410-706-8650;fax: +1-410-706-0346.

E-mail address: [email protected] (H. Ghandehari).1 Present address: Department of Bioengineering, University of

Washington, P.O. Box 352255, AERL 341A, Seattle, WA 98195,USA.

2 Present address: Shire Laboratories Inc., Department of Prefor-mulation Sciences, 1550 East Gude Drive, Rockville, MD 20850,USA.

El-Sayed et al., 2002, 2003). Our previous studieson the transport of cationic PAMAM-NH2 den-drimers of generations 0–4 (G0–G4) showed thatthe smaller dendrimers, G0–G2, exhibited mod-erate permeability across Caco-2 cell monolayers(1×10−6–10×10−6 cm/s) (El-Sayed et al., 2002). Ofthese three dendrimers, G2 appeared to be the mostpromising candidate for oral delivery due to its higherpermeability, as well as its higher number of surfaceamine groups that can be used for immobilization ofdrug molecules. The objective of this research was aninitial attempt to investigate the mechanism(s) of G2transport across Caco-2 cell monolayers.

0378-5173/$ – see front matter © 2003 Elsevier B.V. All rights reserved.doi:10.1016/S0378-5173(03)00391-0

Page 2: Transport mechanism(s) of poly (amidoamine) dendrimers across Caco-2 cell monolayers

152 M. El-Sayed et al. / International Journal of Pharmaceutics 265 (2003) 151–157

PAMAM-NH2 (G2) (Sigma-Aldrich Co., St. Louis,MO) was labeled using fluorescein isothiocyanate(FITC, Sigma-Aldrich Co., St. Louis, MO) followinga previously reported method (El-Sayed et al., 2002).Caco-2 cell monolayers were prepared by standardprocedures described previously and used for trans-port experiments 21–28 days after seeding (El-Sayedet al., 2002).

Permeability of FITC-labeled G2 across Caco-2cell monolayers was investigated in triplicate at adonor concentration of 10.0 mM at pH 7.0, 37◦C,5% CO2, 95% relative humidity, and shaking at 50revolutions per minute (RPM) while maintaining sinkconditions. The contribution of adsorptive endocyto-sis was investigated by determining G2 permeabilityat a lower temperature of 4◦C. The permeability of14C-mannitol as control was not affected by temper-ature. It is assumed that the lower temperature willreduce the contribution of energy-dependent mecha-nisms such as adsorptive endocytosis. Contribution ofthe P-gp efflux system to G2 transport was also ex-amined by determining the AB and BA permeabilityof 14C-paclitaxel at a donor concentration of 200 nM(American Radiolabeled Chemicals, Inc., St. Louis,MO) in presence of G2 (10.0 mM) and by determiningAB and BA permeability of G2 (10.0 mM) in presenceof paclitaxel (200 nM) under similar experimentalconditions.

The permeability of G2 and14C-mannitol wasinvestigated in the presence of palmitoyl carnitine(0.3 mM) to determine the contribution of the para-cellular pathway to G2 transport (Knipp et al., 1997).Permeability samples were collected from the receivercompartment at 30, 90, and 150 min. The 30-minsample was discarded to ensure steady state transportand sink conditions were maintained throughout theexperiments. G2 permeability samples were analyzedusing an HPLC system (Hewlett Packard, Palo Alto,CA) with fluorescence detection (Shimadzu Corpo-ration, Kyoto, Japan) following a method reportedearlier (El-Sayed et al., 2002). Permeability samplesof radioactive nuclides were analyzed by liquid scin-tillation counting (Beckman Coulter, Fullerton, CA).The permeability of Caco-2 cell monolayers (Pm) ofG2 and various radioactive nuclides was determinedfollowing an equation reported earlier (El-Sayed et al.,2002). t-test (Excel®) was used to establish statisticalsignificance between different groups (P < 0.05).

AB permeability of G2 (10.0 mM), across Caco-2cell monolayers, was investigated at 90 and 150 minat both 37 and 4◦C (Fig. 1). Permeability of G2 at37◦C increased with the increase in incubation timethat is similar to the trend reported earlier (El-Sayedet al., 2002). Permeability of G2 at 4◦C was sig-nificantly lower (P < 0.05) than that observed at37◦C at both incubation times of 90 and 150 min.The observed decrease in G2 permeability at 4◦Cin comparison to that observed at 37◦C suggeststhe contribution of adsorptive endocytosis to G2transport across Caco-2 cell monolayers. It mustbe noted however, that this observation is sugges-tive since temperature changes may also result inchanges in the diffusivity of the probes, membranefluidity, transporter activity, intracellular traffickingand other factors that may contribute to changes inpermeability.

The permeability of14C-paclitaxel, a well-knownP-gp substrate, across Caco-2 cell monolayers was de-termined in both AB and BA directions in the presenceof G2 (10.0 mM) at incubation times 90 and 150 min.Results at both time points were similar, therefore,only the results at the last time point (150 min) are re-ported (Fig. 2). In the control experiment, BA perme-ability of 14C-paclitaxel was four-fold higher than thecorresponding AB permeability demonstrating a func-tioning P-gp efflux system in the Caco-2 cell mono-layers used in this study (Fig. 2A). Results indicatethat AB and BA permeability of14C-paclitaxel did notchange in the presence of G2 (10.0 mM) (P < 0.05)(Fig. 2A).

The BA permeability of G2 (10.0 mM) was higherthan the corresponding AB permeability (Fig. 2B)and is similar to the profile reported earlier (El-Sayedet al., 2002). This suggested the contribution of pos-sibly one or more of the efflux systems located atthe apical surface of Caco-2 cells, such as P-gp andABC transporters, to the observed G2 permeabilityprofile. This prompted us to examine the potentialcontribution of P-gp efflux system on the directionalpermeability of these cationic dendrimers. Both ABand BA permeability of G2 did not change in thepresence of paclitaxel (200 nM) suggesting that G2permeability does not involve the P-gp efflux system.The observed directional permeability of G2 appearsto be due to the reported differences in the struc-ture of the tight junctions of Caco-2 cell monolayers

Page 3: Transport mechanism(s) of poly (amidoamine) dendrimers across Caco-2 cell monolayers

M. El-Sayed et al. / International Journal of Pharmaceutics 265 (2003) 151–157 153

0.16 0.11

4.21

2.37

0

1

2

3

4

5

6

90 min 150 min

Per

mea

bili

ty x

10

6 (cm

/sec

)

* *

Fig. 1. AB permeability of G2 across Caco-2 cell monolayers at incubation times of 90 and 150 min at 37◦C (�) and 4◦C (�). Statisticaldifference atP < 0.05 is denoted by∗. Results are reported as mean± standard error of the mean.

where tight junctional proteins are present only at theapical but not at the basolateral membrane (Noachet al., 1993). However, it must be noted that sinceG2 at 10 mM causes the opening of tight junctions,this may result in a change in the activity of thecells with respect to efflux transport mechanisms.Moreover, transporters other than P-gp can poten-tially be involved in PAMAM efflux. Therefore, thecontribution of efflux systems cannot be completelyruled out at this time and further experiments areneeded.

The AB permeability of G2 increased signif-icantly (P < 0.05) from 3.21 × 10−6 cm/s to30.71 × 10−6 cm/s in the presence of palmitoyl car-nitine at an incubation time point of 150 min (Fig. 3).Under similar experimental conditions, palmitoylcarnitine also caused a statistically significant (P <

0.05) increase in14C-mannitol AB permeability from1.09× 10−6 cm/s to 35.43× 10−6 cm/s (Fig. 3). Theincrease in G2 and14C-mannitol AB permeabilityvalues was similar in the presence of palmitoyl car-nitine suggesting the involvement of the paracellularpathway to G2 transport across Caco-2 cell mono-

layers. This conclusion is further supported by theobserved incubation time-dependent decline in TEERvalues (Fig. 4A) and the corresponding increase in14C-mannitol permeability (Fig. 4B) upon apicaland basolateral incubation of G2 (10.0 mM) withCaco-2 cell monolayers. These results suggest that G2modulates the tight junctions of Caco-2 cell mono-layers, thus causing an increase in the paracellularpermeability.

Based on the macromolecular nature of PAMAMdendrimers in general, and the compact sphericalgeometry, hydrophilicity and positive charge of G2in particular, we hypothesized that transport of G2across Caco-2 cell monolayers may be attributed toa combination of transcellular transport by adsorptiveendocytosis and paracellular transport via the tightjunctions. G2 exhibited lower permeability at 4◦Ccompared to its permeability at 37◦C at a given incu-bation time point suggesting it could be transportedin part by adsorptive endocytosis. It is important tonote that reduced permeability at lower temperaturecan also be attributed to inhibition of carrier-mediatedtransport mechanisms, however the transport of G2

Page 4: Transport mechanism(s) of poly (amidoamine) dendrimers across Caco-2 cell monolayers

154 M. El-Sayed et al. / International Journal of Pharmaceutics 265 (2003) 151–157

(A)

8.20

2.01

10.32

3.22

0

2

4

6

8

10

12

AB BA

Per

mea

bilit

y x

106 (c

m/s

ec)

(B)

3.21

10.94

2.94

11.15

0

2

4

6

8

10

12

14

AB BA

Per

mea

bili

ty x

10

6 (cm

/sec

)

Fig. 2. (A) AB and BA permeability of14C-paclitaxel across Caco-2 cell monolayers in absence (�) and presence (�) of G2 (10.0 mM)at incubation time of 150 min. (B) AB and BA permeability of G2 (10.0 mM) across Caco-2 cell monolayers in absence (�) and presence(�) of paclitaxel (200 nM) at incubation time of 150 min. Results are reported as mean± standard error of the mean.

Page 5: Transport mechanism(s) of poly (amidoamine) dendrimers across Caco-2 cell monolayers

M. El-Sayed et al. / International Journal of Pharmaceutics 265 (2003) 151–157 155

1.00

3.21

35.43

30.71

0

5

10

15

20

25

30

35

40

45

G2 Mannitol

Per

mea

bili

ty x

10

6 (cm

/sec

)

Fig. 3. AB permeability of G2 and14C-mannitol across Caco-2 cell monolayers in absence (�) and presence (�) of 0.3 mM palmitoylcarnitine. Results are reported as mean± standard error of the mean.

is unlikely to involve such systems due to their largemacromolecular nature and lack of structural simi-larity to the established substrates for such carriersystems.

Our previous studies showed that the BA perme-ability of G2 was higher than the corresponding ABpermeability under the same experimental conditions(El-Sayed et al., 2002) which suggested the contribu-tion of an efflux system. Based on the results of thisstudy, it appears that the observed directional perme-ability of G2 is due to the differences in the struc-ture of the tight junctions of Caco-2 cell monolayerswhere tight junctional proteins are present at the api-cal but not at the basolateral membranes (Noach et al.,1993).

The hydrophilic and cationic nature of G2 (re-sulting in modulation of the tight junctions) coupledwith its compact spherical geometry in solution sug-gest that G2 may also permeate across Caco-2 cellmonolayers via the paracellular route. Applicationof permeation enhancers such as palmitoyl carnitinehas been shown to increase Caco-2 permeability formolecules that are transported via the paracellularroute by dilation of the paracellular space (Knipp

et al., 1997). Our results show that palmitoyl car-nitine caused an increase in AB permeability ofG2 that was similar to the observed increase in14C-mannitol permeability, a known paracellular per-meability marker (Fig. 3). This suggests that G2 is inpart transported via the paracellular route. This ob-servation is further supported by the time-dependentdecline in TEER and the corresponding increasein permeability of14C-mannitol across Caco-2 cellmonolayers upon incubation with G2 (10.0 mM)(Fig. 4).

In conclusion, our initial observations suggest thatthe permeability of G2 across Caco-2 cell monolayersoccurs by a combination of paracellular transport andan energy-dependent process such as adsorptive en-docytosis. In addition, G2 dendrimers do not appearto be substrates for the P-gp efflux system. Whilethese observations elucidate potential mechanismsof PAMAM transport across intestinal epithelial bar-riers, additional studies allowing the quantificationand visualization of paracellular and transcellularcomponents are needed to further clarify the mech-anisms of transport of these macromolecular drugcarriers.

Page 6: Transport mechanism(s) of poly (amidoamine) dendrimers across Caco-2 cell monolayers

156 M. El-Sayed et al. / International Journal of Pharmaceutics 265 (2003) 151–157

(A) 0

20

40

60

80

100

120

0 30 60 90 120 150

Time (min)

% D

eclin

e in

TE

ER

(B) 0

5

10

15

20

25

30

30 60 90 120 150

Time (min)

Per

mea

bili

ty x

10

6 (cm

/sec

)

Fig. 4. Effect of apical (�) and basolateral (�) incubation of G2 on: (A) transepithelial electrical resistance (TEER) and (B)14C-mannitolpermeability across Caco-2 cell monolayers. Results are reported as mean± standard error of the mean.

Page 7: Transport mechanism(s) of poly (amidoamine) dendrimers across Caco-2 cell monolayers

M. El-Sayed et al. / International Journal of Pharmaceutics 265 (2003) 151–157 157

References

El-Sayed, M., Ginski, M., Rhodes, C.A., Ghandehari, H., 2002.Transepithelial transport of poly (amidoamine) dendrimers ac-ross Caco-2 cell monolayers. J. Control. Release 81, 355–365.

El-Sayed, M., Ginski, M., Rhodes, C.A., Ghandehari, H.,2003. Influence of surface chemistry of poly (amidoamine)dendrimers on Caco-2 cell monolayers. J. Bioactive CompatiblePolym. 18, 93–111.

Knipp, G.T., Ho, N.F.H., Barsuhn, C.L., Borchardt, R.R., 1997.Paracellular diffusion in Caco-2 cell monolayers: effect ofperturbation on the transport of hydrophilic compounds thatvary in charge and size. J. Pharm. Sci. 86, 1105–1110.

Noach, A.B.J., Kurosaki, Y., Blom-Roosemalen, M.C.M., deBoer, A.G., Breimer, D.D., 1993. Cell-polarity dependenteffect of chelation on the paracellular permeability ofconfluent Caco-2 cell monolayers. Int. J. Pharm. 90, 229–237.

Wiwattanapatapee, R., Carreno-Gomez, B., Malik, N., Duncan, R.,1998. PAMAM dendrimers as a potential oral drug deliverysystem: uptake by everted rat intestinal sacs in-vitro. J. Pharm.Pharamcol. 50, 99.

Wiwattanapatapee, R., Carreno-Gomez, B., Malik, N., Duncan,R., 2000. Anionic PAMAM dendrimers rapidly cross adult ratintestine in vitro: a potential oral delivery system? Pharm. Res.17, 991–998.